ABSTRACT
A comprehensive diagnostic assessment is needed to improve understanding of the health status of patients with chronic obstructive pulmonary disease (COPD) or asthma. Therefore, this study investigated which components and subsequent instruments should be part of a holistic assessment in secondary care. We also explored which data need to be exchanged for an adequate transfer of patients between primary and secondary care, and vice versa. A cross-sectional Web-based survey was conducted among Dutch healthcare professionals using a Delphi-like procedure; these included professionals working in primary or secondary care, medical advisors of health insurance companies and patients' representatives. The national guidelines were used as a starting point, resulting in a questionnaire addressing 55 components related to a comprehensive diagnostic assessment, covering the domains physiological impairments, symptoms, functional limitations and quality of life. Of the 151 experts and stakeholders invited, 92 (60.9%) completed the first round and 79 (52.3%) the second round; most respondents were pulmonologists. There was a high level of agreement between respondents from primary versus secondary care regarding which components should be measured during a comprehensive assessment of patients with asthma or COPD in secondary care and the instruments to measure these components. Regarding the exchange of information, upon referral, pulmonologists required little information from the general practitioners, whereas general practitioners required more extensive information after referral. An overview is provided of what should be part of a holistic assessment of health status in asthma and COPD. This information can be used as input for integrated care pathways.
Background
Chronic obstructive pulmonary disease (COPD) is a complex and heterogeneous disease with pulmonary and systemic manifestations and multiple factors that affect a patients' health status as defined by four domains: physiological impairments, symptoms, functional limitations and quality of life Citation(1–3). An assessment of all four domains might improve understanding of the burden of disease with subsequent guidance by treatable traits Citation(4). A similar line of reasoning applies to patients with asthma, where multiple factors (e.g., airflow obstruction with dyspnoea and cough, bronchial hyperactivity, adherence to medication, overweight, a low level of physical activity, functional deficits and mood) affect the burden of disease Citation(5). Assessing only one single factor (or a selection of these factors) might lead to an underestimation of the severity, incomplete management and disappointing outcome of care.
There is an increasing interest in the existence of a mixed COPD-asthma phenotype, known as the asthma-COPD overlap syndrome (ACOS). Although they have different characteristics, some individuals share features of both diseases Citation(6,7). At first presentation, it may be difficult to distinguish between COPD, asthma and an overlap of both diseases. In particular, patients with more advanced or more complex COPD and/or asthma might benefit from a more structured approach towards assessment of all four domains of health status Citation(8).
Although the relevance of each of these domains is acknowledged, there is little consensus on which components have to be addressed during routine assessment by pulmonologists Citation(3). Furthermore, it is unclear which type of information needs to be exchanged between the general practitioner (GP) and the pulmonologist upon referral and on return to the GP after in-depth assessment for further treatment Citation(9).
Therefore, the aims of this study are twofold. First, to determine which components experts consider being important in the comprehensive diagnostic assessment of the health status of patients with asthma and COPD by the pulmonologist. Second, to determine which data should be exchanged between the GP and the pulmonologist upon referral, and on return to the GP.
In addition, we investigated whether opinions on these aspects differed between primary care (PC) and secondary care (SC), as well as between pulmonologists and respiratory nurses.
Methods
Study design
A cross-sectional Web-based survey was conducted among Dutch expert healthcare professionals using a Delphi-like procedure with two rounds, between February and October 2014 Citation(10–12). First, we asked which components should be part of the comprehensive diagnostic assessment, and in the exchange of information between PC and SC. Second, we asked the expert panel how these components should (preferably) be measured.
Recruitment of experts
The sample comprised dedicated experts and stakeholders in Dutch healthcare with known interest in and knowledge of asthma and COPD. For this, we contacted the professional respiratory associations working groups. These working groups are responsible for the implementation of innovations, and they are mandated by the professionals to decide on the content of the guidelines in the Dutch healthcare system; national issues like this nationwide Delphi study will therefore always be referred to these working groups. This resulted in a panel of 151 experts and stakeholders, recommended by their respective professional associations. These included pulmonologists participating in the working group on asthma or COPD from the Dutch Society of Pulmonologists (in Dutch: NVALT), respiratory nurses working in SC and recommended by the Dutch Nursing Association, GPs who were members of the Dutch Organisation of GPs with special interest in asthma and COPD (in Dutch: CAHAG), medical advisors of health insurance companies (Achmea, Menzis, VGZ) and representatives of the Lung Foundation Netherlands.
Questionnaire round 1
The first round of the survey was conducted in February 2014. Part of the questionnaire was a list of 40 potentially important components of a comprehensive diagnostic assessment. The 40 components were obtained by a two-stepped method. The Dutch Standards of Care for patients with asthma and COPD were starting point of the list with components. These standards are developed by a working group in which experts of all disciplines are represented. The development is based on literature, from which the components of a diagnostic assessment were extracted and grouped into four domains: physiological impairment, symptoms, functional impairment and health-related quality of life (HRQoL). Based on these standards, a comprehensive and detailed list supplemented with components used in clinical care was presented to six independent Dutch key-opinion leaders. In the final list, their recommendations were included.
These components were allocated to the four domains of health: physiological impairments, symptoms, functional limitations and quality of life Citation(3). Another 15 components grouped into the category ‘interpretation of data’ were added to explore to what extent the experts would value data and an explanation of the measurements and other relevant information, in order to complete a holistic assessment and establish an appropriate care plan.
Experts were asked to rate the importance of each of the 55 components using a 7-point Likert scale (range: 1 = very important; 7 = very unimportant). Next, the experts were asked to indicate which information about each component should be exchanged between PC and SC.
Examples of the questions are:
| 1) | How important is this component in the comprehensive diagnostic assessment of asthma or COPD in SC? | ||||
| 2) | How important is it to report this component during transfer from PC to SC for a comprehensive diagnostic assessment? | ||||
| 3) | How important is it to report this component during transfer from SC back to PC for further treatment or shared care? | ||||
All three questions were asked separately for each component. Several components were specific for asthma or COPD, which was indicated for each specific question. After each question, experts were invited to add comments in an open space. In addition, experts were asked to select and rank the eight most important items. The responses obtained with questionnaire 1 were analysed in June 2014 and were used to design the questionnaire of round 2.
Questionnaire round 2
The second round was conducted in October 2014. The aim of the second round was to define which type of information is required for each component and how that information should be measured (i.e., which instrument, test, tool or questionnaire should be used). If ≤75% of the experts rated a component as important in the first round, it was excluded from the second round.
Two specific questions were addressed in the survey:
| 1) | What would you like to know about the specific component? | ||||
| 2) | Which measurement instrument(s) is (are) most appropriate? | ||||
For round 2, we approached the same experts as in round 1. Experts received a list of pre-selected options (obtained from clinical guidelines, literature and expert opinion) to choose from and were allowed to choose multiple items. In addition, space was provided after each question to add comments.
Data collection
All respondents received an email with an invitation letter and a link to a Web-based questionnaire: ThesisTool for the first round (www.thesistools.be) and SurveyMonkey for the second round (https://nl.surveymonkey.com). Email reminders were sent after 4 weeks. Participation in the study was voluntary. Experts and stakeholders participated anonymously to avoid the authority, personality or reputation of some participants from dominating others in the study process. To some extent, it might also free participants of their personal biases, allow free opinions and encourage open critique.
In both rounds, respondent's socio-demographic data such as age, gender, profession, employed in PC/SC and years of work experience were also collected.
Data analysis
In the first round, a component was defined as relevant when ≥75% of the respondents chose either ‘important’ or ‘very important’ (Likert scale 1–2). Binary logistic regression analysis with ‘component selected as important’ as dependent variable and ‘age’ and ‘gender’ as co-variables was used to assess whether there was a difference in importance ratings between professionals working in PC and in SC. Binary logistic regression was also performed to investigate whether there was a difference in importance ratings between pulmonologists and respiratory nurses. The results were expressed as odds ratios (OR) and 95% confidence interval. Results of the second round were summarised with descriptive statistics (n, percentage by professional).
Results
Respondents' characteristics
Of the 151 participating experts and stakeholders, 92 (60.9%) completed the first round and 79 (52.3%) the second round. Most of the respondents in both rounds were pulmonologist (53% and 57%, respectively). In the first round, 27% of the respondents were respiratory nurses and 17% GPs. In the second round, 24% of the respondents were respiratory nurses and 14% GPs. summarises the respondents' characteristics. Most of the respondents were working in a hospital setting and had ≥5 years of work experience ().
Table 1. Characteristics of the respondents.
Results of the first round
How important is the component in the comprehensive diagnostic assessment in secondary care?
shows the importance ratings of the 55 components, which are grouped into the four health domains and the additional interpretation category (middle column). 39 (71%) of the 55 components were rated as important by ≥75% of the respondents. In particular, almost all the components in the domains functional limitation and quality of life were considered important. Most of the components found to be less important (i.e., <75% of respondents rated this as important) were in the physiological impairment domain or referred to disease categorization (e.g., asthma phenotype or GOLD A–D) in ‘Interpretation of data’ (). Three components, ‘peak flow measurement’, ‘inflammatory assessment’ and ‘vocal symptoms’ were considered unimportant because only 30% of the respondents rated them as important.
Table 2. Components to be measured in a secondary care assessment.
The right-hand side of shows the difference in scoring between PC and SC. The adjusted ORs ≥1 show that respondents from PC were more likely to rate components of the functional limitation domain and the ‘Interpretation of data’ category as being important than respondents from SC. Significant ORs were found for ‘family history’, ‘routine laboratory’, ‘muscle strength’ and ‘phenotyping COPD’. No components were found to be significantly more important by respondents from PC or from SC.
Furthermore, respiratory nurses were more likely to rate ‘quality of life’ and ‘family history’ as more important than pulmonologists (data not shown).
What components are important to exchange between primary care and secondary care, and vice versa?
(second column) shows the components considered necessary for an adequate documentation of a patient at referral from PC to SC. Ten components, mostly from the domain physiological impairment (8/10), were considered important for an adequate referral from PC to SC by ≥75% of the respondents.
Table 3. Exchange of information.
(fourth column) shows that PC professionals rated many more components as being important (indicated by an OR >1) to report at referral to SC than respondents from SC. Again, no components for referral were rated more important by PC or SC professionals. More items were considered relevant for transfer from SC to PC for further treatment by the GP or shared care. (fourth column) shows that, of the 55 components, 38 (69%) were marked as important by at least 75% of the respondents. The components ‘peak flow measurement’ and ‘vocal symptoms’ were considered irrelevant (<30%).
Again, respondents from PC rated more components as being more important than respondents from SC (, fourth column).
Results of the second round
We excluded the PC in the second round because of the low response rate.
shows the preferred instruments to measure the components as selected by the SC professionals (n = 64, 81%; of which 45 pulmonologists and 19 respiratory nurses). Measurements chosen by <50% of the respondents are presented in the online supplement.
Table 4. How would secondary care like to measure this component?
In general, there was a high preference for registering information as open format in the medical files. If a standardised measurement was chosen, short questionnaires were preferred, like the Medical Research Council (MRC), Asthma Control question (ACQ) and the Clinical COPD question (CCQ). Overall, there was a high level of agreement between pulmonologists and respiratory nurses. Pulmonologists gave less priority to more specific nursing skills, such as inhaler device handling and nutritional assessment.
Discussion/conclusion
Our multidisciplinary expert panel showed an overall high level of agreement among respondents about the components that should be measured during a comprehensive assessment of patients with asthma or COPD in SC. Moreover, both PC and SC professionals had similar opinions regarding the components to be measured. Less agreement was found between PC and SC with regard to the exchange of information on referral, especially regarding the components on referral from PC to SC. When the expert panel was asked how each component should be measured by questioning if applicable, they showed a preference for short questionnaires.
Agusti and MacNee Citation(4) suggested to implement a ‘control panel’ that includes three different domains of the disease to optimise the assessment of individual patients with complex diseases as COPD, whereas Vercoulen et al. Citation(3,13,14) argued that a detailed assessment needs to include four domains. The Dutch Standards of Care for patients with asthma or COPD were built on these examples and adopted a model in which the assessment is based on four domains: physiological impairment, symptoms, functional impairment and quality of life (including HRQoL) Citation(15,16). Such an assessment requires a combination of several instruments. However, there is relatively little guidance on the exact content and corresponding instruments to measure these domains. This observation was the main reason to conduct the current study which, to our knowledge, is the first that systematically asks experts from both PC and SC what components should be measured to cover the four domains, and which instruments are most useful to measure them. In contrast to other studies, we added a category ‘Interpretation of data’. This final step integrates the available information into a truly holistic evaluation of the patient's health status with subsequent therapeutic implications. All earlier publications on holistic assessments lack this final and essential step based on the data of the four domains Citation(2–4,14).
In general, respondents agree on the components that should be measured in the assessment. These responses are in line with the recommendations currently made in the Dutch standards of care for Asthma or COPD but require even more extensive information Citation(15,16). This is also consistent with the elements of the recently developed Dutch Assessment of Burden of COPD (ABC) tool, which visualises multiple different aspects of the burden of COPD, thereby facilitating shared decision-making Citation(17).
In particular, several items that reflect (lack of) consensus seem remarkable and are discussed below by category.
Regarding the physiological impairments, there is ongoing debate concerning the role of high-resolution computed tomography as part of routine diagnosis in COPD and asthma. Gupta et al. Citation(18) recommend (HR)CT in all patients with asthma to detect common abnormalities (particularly bronchiectasis), and Hardin et al. Citation(19) promote (HR)CT in all patients to better characterise the clinical features of the asthma-COPD overlap group. However, only 34% of our respondents endorse this opinion. Although (HR)CT provides unique COPD phenotyping information that is potentially predictive of exacerbations in subgroups of patients, the clinical importance with subsequent treatable traits of the (HR)CT in asthma and COPD is still debated Citation(20,21).
This also applies to measurement of the diffusion capacity for carbon monoxide (DL, CO). A vast majority of the respondents (83%) designated DL, CO as important. Indeed, DL, CO is an excellent test to unravel the physiological disorder in severe COPD. An impaired DL, CO can be used as additional argument for co-existing COPD in smokers with asthma. Hence, the DL, CO measurement has relevant diagnostic but little therapeutic consequences Citation(22,23). This is in contrast with airflow obstruction and (dynamic) hyperinflation, which may define important targets for medical treatment in both asthma and COPD.
Regarding the symptoms domain, the component fatigue was selected by 66% of the respondents as important, which is below our threshold of priority. Reason for this low response is probably because the factors that contribute to fatigue are diverse and perhaps poorly understood, so that the impact of treatment on fatigue is viewed as limited. However, fatigue is a disruptive symptom that inhibits normal functional performance in daily activities and is ranked as an important symptom by patients. This is why the Netherlands Respiratory Society included fatigue as one of the five themes in the National Program Lung Research. Furthermore, fatigue is also an item in the Assessment of Burden of COPD (ABC) Tool. Earlier studies proposed to integrate the Functional Assessment of Chronic Illness Therapy scale for fatigue (FACIT-F) in the assessment Citation(24).
Regarding the domain functional limitation, patients with asthma, and especially COPD, have impaired yet subjectively overestimated performance of physical activity that seems to be related to their impaired health status Citation(25,26). Despite this, monitoring physical activity is not yet commonly included in patients’ assessments during routine care. The scores indicating high importance for physical activity and exercise capacity support their clinical importance and suggest increased awareness in both PC and SC. Respondents prefer an accelerometer as objective measurement tool. We suggest that objective physical activity measurement strategies and interventions aiming to increase physical activity should be implemented Citation(25,27).
Regarding the category ‘Interpretation of data’ of measurements, disease categorisation, and more specifically, phenotyping in asthma versus specifying the GOLD ABCD classification in COPD, has no particular interest for the respondents. The first could be explained by the fact that phenotyping as first step in the assessment of asthma might not have direct added value in the treatment. However, the importance of phenotyping in difficult asthma has been reported Citation(8). The lack of interest for the ABCD GOLD strategy in COPD might be explained by the fact that the consequences of the treatment recommendations as part of the ABCD strategy have not yet been validated in the clinical context.
When asking what information should be exchanged upon referral, pulmonologists generally require little information from the GP, i.e., even less than the GP is willing to provide. This is probably because the pulmonologist will in any case repeat the assessment, especially in case of lung function measurements: only 31% of the SC professionals will use the spirometry performed in PC. This might be due to perceived quality issues Citation(28). In addition, pulmonologists probably prefer to measure spirometry as part of a more extensive lung function test in their own laboratory.
In contrast, the GP demands extensive information on all four domains of health status when patients are referred back to PC. The need for more information in this situation suggests awareness of multiple health status indicators and their implications for further treatment in PC. It also indicates that the GP judges the pulmonologist to be the designated expert to deliver this information. The need for extensive information also requires an adequate and standardised exchange of information. Inadequate information exchange might be a reason for poor outcomes in the management of chronic disorders Citation(9). Han et al. Citation(21) have pleaded for smooth and adequate communication between healthcare providers, stating that truly integrated care allows the patient to transition smoothly and seamlessly between health providers.
In general, there was a high level of agreement on the instrument and tools to collect the required information. There is high preference for registering information as open format in the medical records. Although this may seem easy, it hampers uniformity of registration and subsequent treatment. Co-morbidity potentiates the morbidity of COPD, increases the risk of hospitalisation and healthcare costs, and needs to be evaluated to develop treatment guidelines to improve benefits for the individual patient Citation(29). A more structured registration by the Charlson Comorbidity Index, or clustering in five identified clusters, has been recommended Citation(30,31). In addition, registration of HRQoL might benefit from structured, transparent, interchangeable registration, in this respect with validated questionnaires. Short validated instruments (like the CCQ, ACQ and MRC) are preferred to the more extensive but internationally accepted instruments like the Saint George's Respiratory Questionnaire. Short instruments are easy to use and implement in daily clinical practice and provide direct feedback Citation(32).
The present study has several strengths and limitations.
The Dutch Standards of Care for asthma and COPD recommend a productive interaction between SC and PC in which the diagnostic secondary care assessment adds to the diagnostic assessment in primary care. Due to this integrated approach, we decided to include respondents from PC and SC.
While this subject is innovative and in the Dutch healthcare system still in development, the sample of respondents consists of specifically dedicated disciplines from PC and SC with adequate participation of specific asthma and COPD working groups. These working groups are responsible for the implementation of innovations, and they are mandated by the professionals to decide on the content of the guidelines in the Dutch healthcare system; national issues like this nationwide Delphi study will therefore always be referred to these working groups. This is the most common and accepted way in the Netherlands to address disease-specific issues and has several advantages; knowledge is concentrated and up-to-date, and it may ensure a higher commitment if conclusions from this study regarding the measurements are to be implemented in daily practice. The disadvantage of this selection is the probability of a higher consensus than among a random sample of GPs.
It should be noted that SC was relatively over-represented in round 2 of the Delphi procedure. The lower number of GP responses in round 2 is probably due to the specific nature of the questions, addressing diagnostic issues in SC. However, as round 2 asked about tools that should be used during the assessment in SC, it is not a major shortcoming.
As in many studies, our concern was adequate response. Overall, the response in the first round was 60.9%, and that in the second round was 52%. This is in line with other Delphi procedures Citation(33,34). However, response among the individual disciplines is low. Several studies argue that a sample size of 9 to maximally 13 participants is sufficient because inclusion of more participants does not generate new insights, i.e., saturation is achieved Citation(12,35). In the first round, 15 GPs (44%) responded, and in the second round, only 8 responded. Therefore, we decide to exclude the GPs in the second round. This is shown in
We used a modified version of the Delphi procedure, with a cut-off of 75% scoring of 6 or 7 on a 7-point Likert scale in order to define a component as important. Although this cut-off point is somewhat arbitrary, it is not unrealistic compared with that of other studies Citation(36,37). Moreover, cut-offs vary widely between different Delphi procedures.
The optimal care for patients with asthma and COPD requires an individualised approach that recognises all aspects of the diseases and commitment from all the stakeholders. Although several integrated care programs, which apply such an individualised approach, are available, they all lack a detailed specification of instruments used in the holistic assessment Citation(3,38,39). The present study provides more insight into the preferred components of such an assessment and the instruments used to measure them Citation(38). This information might also be used as input for shared care programs, the development of Information and Communication Technology (ICT) systems that support the standardised recording of the assessment results, and perhaps to support reimbursement negotiations with healthcare insurance companies. The components for which no clear consensus is reached () should not be ignored but need to be analysed further.
Declaration of interest
The authors report no conflicts of interest.
Funding
The authors wish to thank “PICASSO for COPD” and AstraZeneca for funding this study.
References
- Nussbaumer-Ochsner Y, Rabe KF. Systemic manifestations of COPD. Chest 2011; 139(1):165–173.
- Wilson IB, Cleary PD. Linking clinical variables with health-related quality of life. A conceptual model of patient outcomes. JAMA 1995; 273(1):59–65.
- Vercoulen JH, Daudey L, Molema J, Vos PJ, Peters JB, Top M, et al. An Integral assessment framework of health status in chronic obstructive pulmonary disease (COPD). Int J Behav Med 2008; 15(4):263–279.
- Agusti A, MacNee W. The COPD control panel: towards personalised medicine in COPD. Thorax 2013; 68(7):687–690.
- Wechsler ME. Managing asthma in primary care: putting new guideline recommendations into context. Mayo Clin Proc 2009; 84(8):707–717.
- Barrecheguren M, Esquinas C, Miravitlles M. How can we identify patients with asthma-COPD overlap syndrome in clinical practice? Arch Bronconeumol 2016; 52(2):59–60.
- Barrecheguren M, Esquinas C, Miravitlles M. The asthma-chronic obstructive pulmonary disease overlap syndrome (ACOS): opportunities and challenges. Curr Opin Pulm Med 2015; 21(1):74–79.
- Global Strategy for Asthma Management and Prevention. 2016. Available from: http://ginasthmaorg/2016-gina-report-global-strategy-for-asthma-management-and-prevention/.
- Bodenheimer T. Coordinating care–a perilous journey through the health care system. New Engl J Med 2008; 358(10):1064–1071.
- Boulkedid R, Abdoul H, Loustau M, Sibony O, Alberti C. Using and reporting the Delphi method for selecting healthcare quality indicators: a systematic review. PloS One 2011; 6(6):e20476.
- Steurer J. The Delphi method: an efficient procedure to generate knowledge. Skeletal Radiol 2011; 40(8):959–961.
- Fitch K, Bernstein SJ, Aguilar M, Burnand B, LaCalle J, Lazaro P. The RAND/UCLA appropriateness method user's manual. Available from: http://wwwrandorg/publications/MR/MR1269/Accessed 2001 (Cited 2006).
- Peters JB, Daudey L, Heijdra YF, Molema J, Dekhuijzen PN, Vercoulen JH. Development of a battery of instruments for detailed measurement of health status in patients with COPD in routine care: the Nijmegen Clinical Screening Instrument. Qual Life Res 2009; 18(7):901–912.
- Vercoulen JH. A simple method to enable patient-tailored treatment and to motivate the patient to change behaviour. Chron Respir Dis 2012; 9(4):259–268.
- Nederland LA. Zorgstandaard COPD (Care Standard COPD). Long Alliantie Nederland, 2010.
- Nederland LA. Zorgstandaard Astma volwassenen. Long Alliantie Nederland, 2012.
- Slok AH, In't Veen JC, Chavannes NH, van der Molen T, Rutten-van Molken MP, Kerstjens HA, et al. Development of the Assessment of Burden of COPD tool: an integrated tool to measure the burden of COPD. NPJ Prim Care Respir Med 2014; 24:14021.
- Gupta S, Siddiqui S, Haldar P, Raj JV, Entwisle JJ, Wardlaw AJ, et al. Qualitative analysis of high-resolution CT scans in severe asthma. Chest 2009; 136(6):1521–1528.
- Hardin M, Cho M, McDonald ML, Beaty T, Ramsdell J, Bhatt S, et al. The clinical and genetic features of COPD-asthma overlap syndrome. Eur Respir J 2014; 44(2):341–350.
- Wells JM, Washko GR, Han MK, Abbas N, Nath H, Mamary AJ, et al. Pulmonary arterial enlargement and acute exacerbations of COPD. N Engl J Med 2012; 367(10):913–921.
- Han MK, Martinez CH, Au DH, Bourbeau J, Boyd CM, Branson R, et al. Meeting the challenge of COPD care delivery in the USA: a multiprovider perspective. Lancet Respir Med 2016; 4(6):473–526.
- Miller MR, Crapo R, Hankinson J, Brusasco V, Burgos F, Casaburi R, et al. General considerations for lung function testing. Eur Respir J 2005; 26(1):153–161.
- Pellegrino R, Viegi G, Brusasco V, Crapo RO, Burgos F, Casaburi R, et al. Interpretative strategies for lung function tests. Eur Respir J 2005; 26(5):948–968.
- Antoniu SA, Petrescu E, Stanescu R, Anisie E, Boiculese L. Impact of fatigue in patients with chronic obstructive pulmonary disease: results from an exploratory study. Ther Adv Respir Dis 2016; 10(1):26–33.
- van't Hul AJ, Frouws S, van den Akker E, van Lummel R, Starrenburg-Razenberg A, van Bruggen A, et al. Decreased physical activity in adults with bronchial asthma. Respir Med 2016; 114:72–77.
- Demeyer H, Duenas-Espin I, De Jongh C, Louvaris Z, Hornikx M, Gimeno-Santos E, et al. Can health status questionnaires be used as a measure of physical activity in COPD patients? Eur Respir J 2016; 47(5):1565–1568.
- Spruit MA, Singh SJ, Garvey C, ZuWallack R, Nici L, Rochester C, et al. An official American Thoracic Society/European Respiratory Society statement: key concepts and advances in pulmonary rehabilitation. Am J Respir Crit Care Med 2013; 188(8):e13–e64.
- Enright P. Provide GPs with spirometry, not spirometers. Thorax 2008; 63(5):387–388.
- Barnes PJ, Celli BR. Systemic manifestations and comorbidities of COPD. Eur Respir J 2009; 33(5):1165–1185.
- Jones PW, Watz H, Wouters EF, Cazzola M. COPD: the patient perspective. Int J Chron Obstruct Pulmon Dis 2016; 11 Spec Iss:13–20.
- Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis 1987; 40(5):373–383.
- Slok AH, Bemelmans TC, Kotz D, van der Molen T, Kerstjens HA, In't Veen JC, et al. The assessment of burden of COPD (ABC) scale: A reliable and valid questionnaire. COPD 2016:1–8.
- Camp PG, Reid WD, Chung F, Kirkham A, Brooks D, Goodridge D, et al. Clinical decision-making tool for safe and effective prescription of exercise in acute exacerbations of chronic obstructive pulmonary disease: Results from an interdisciplinary Delphi survey and focus groups. Phys Ther 2015; 95(10):1387–1396.
- Cooke M, Campbell M. Comparing patient and professional views of expected treatment outcomes for chronic obstructive pulmonary disease: a Delphi study identifies possibilities for change in service delivery in England, UK. J Clin Nurs 2014; 23(13–14):1990–2002.
- Francis JJ, Johnston, Marie, Robertson, Clare, Glidewell, Liz, Entwistle, Vikki, Eccles, Martin P., and Grimshaw JM. What is an adequate sample size? Operationalising data saturation for theory-based interview studies. Psychol Health 2009; 25(10):1229–1245.
- Gracht HAvd. Consensus measurement in Delphi studies: review and implications for future quality assurance. Technol Forecasting Soc Change 2012; 79(8):1525–1536.
- JoAnne W. Putman ANS, Bruininks RH. Future directions in educations and inclusion of students with disabilities: a Delphi investigation. Educ Psychol Papers Publications 1995; 61(6):553–576.
- Nici L, ZuWallack R. American Thoracic Society Subcommittee on integrated care of the CP. An official American Thoracic Society workshop report: the integrated care of the COPD patient. Proc Am Thorac Soc 2012; 9(1):9–18.
- Wu CX, Tan WS, See RC, Yu W, Kwek LS, Toh MP, et al. A matched-group study protocol to evaluate the implementation of an integrated care pathway programme for chronic obstructive pulmonary disease in Singapore. BMJ Open 2015; 5(1):e005655.